a survey of converging solutions for heterogeneous mobile...

IEEE WIRELESS COMMUNICATIONS

1

Abstract— In 5th Generation (5G) systems, the current

Machine-to-Machine (M2M) communications using Wi-Fi or

Bluetooth provide a good opportunity to dramatically increase the

overall performance. Converged mobile networks can provide

M2M communications with significant performance improvement

from sharing unlicensed spectrum bands in cellular networks, such

as Long Term Evolution-Advanced (LTE-A), by using cognitive

radio technology. Thus, the converged mobile network will become

one of the most popular future research topics because mobile

multimedia content services have been generally accepted among

mobile device users. In this article, we provide an overview of

converged mobile networks, investigating different types of

converged mobile network, different types of convergence, and the

current problems and solutions. This survey article also proposes

potential research topics in converged mobile networks.

Index Terms: converged mobile networks, interference,

cognitive radio, resource sharing, M2M/LTE-A, 5G

heterogeneous networks, DIDO, unlicensed spectrum

I. INTRODUCTION

INCE the number of mobile multimedia content users has

increased dramatically, a serious problem over limited

spectrum resources has arisen. In order to solve this problem,

research has lately begun into increasing available spectrum

resources by integrating (or converging) several network

resources into a single mobile network. This single converged

mobile network allows each different network to share the idle

resources of other networks. However, each of these existing

networks — as for example cellular networks, video

broadcasting, Wireless Sensor Networks (WSNs), and Wireless

Local Area Networks (WLANs) — works independently using

Manuscript received September 20, 2014. This research was supported by

the National Research Foundation, the Korean government (MEST) (No.

2012-0004811).

Minho Jo is with Korea University, Sejong, S. Korea (Corresponding

author. Phone: +82-44-860-1348; Fax: +82-44-860-1584; e-mail:

[email protected]).

Taras Maksymyuk is with the Korea University, Sejong, S. Korea (Phone:

+82-44-860-1755, e-mail: [email protected]).

Rodrigo L. Batista is with Federal University of Ceará, Fortaleza, Brazil,

(e-mail: [email protected]).

Tarcisio F. Maciel is with Federal University of Ceará, Fortaleza, Brazil,

(e-mail: [email protected]).

André L.F. de Almeida is with Federal University of Ceará, Fortaleza,

Brazil, (e-mail: [email protected]).

Mykhailo Klymash is with, Lviv Polytechnic National University, Lviv,

Ukraine, (e-mail: [email protected]).

its own frequency band. Currently, there already are many

emerging mobile convergence network technologies. Thus, this

survey article addresses the problems and solutions of existing

mobile convergence networks and discusses their future

research topics.

The advent of Long Term Evolution (LTE) technology has

allowed the provision of high-throughput applications for

mobile users. In addition, the development of hardware such as

smartphones/tablets and mobile operating system (OS) such as

Android, iOS, Windows Phone, and Tizen have accelerated the

mobile service development, requiring plenty of spectrum

bands. Indeed, we have seen an extremely wide range of various

mobile applications, web services, and real-time multimedia

services these days. Lately, most of these services have been

available only for fixed-location users.

A popular topic in the current information and

communication technologies area is cloud computing, which is

widely recognized as a next-generation computing

infrastructure. Cloud services enable users to adaptively utilize

resources, and thus, allow service providers to easily implement

mobile applications with minimal management efforts. Mobile

cloud computing brings new types of services and allows mobile

users to access services previously available only to desktop

computer users, including real-time services, interactive games

and ultra-high definition (UHD) video streaming, which

requires much higher performance from mobile networks. Even

LTE networks are not able to properly support them. Therefore,

some new technologies for increasing performance, including

fifth-generation (5G) mobile networks, have been proposed

over the last few years [1]. Convergence of heterogeneous

wireless access techniques has emerged as one of the key

solutions for 5G mobile networks. In mobile converged

networks, users access many kinds of mobile services, despite

their location, connection type and device.

The converged mobile network allows users to effectively

utilize the multiple transmitters/receivers operating in standards

such as Long Term Evolution-Advanced (LTE-A), Wi-Fi,

Bluetooth, ZigBee, etc. For example, users can be reallocated

between different types of network (LTE, LAN, Wi-Fi, ZigBee,

etc.) without session interruption, or even simultaneously utilize

resources from different networks [2].

In this paper, we provide an overview of converged

(heterogeneous) mobile networks and then we discuss key

problems and existing solutions for the coexistence of different

A Survey of Converging Solutions for

Heterogeneous Mobile Networks

Minho Jo, Member, IEEE, Taras Maksymyuk, Member, IEEE, Rodrigo L. Batista, Tarcisio F. Maciel,

André L.F. de Almeida, Senior Member, IEEE and Mykhailo Klymash, Member, IEEE

S


2

wireless protocols, such as cellular, broadcast, Wi-Fi, and

Bluetooth networks, within the same spectrum band. Utilization

of unlicensed spectrum bands in converged mobile networks is

addressed by using cognitive radio technology [3].

This article is organized as follows. In Section II, an overview

of mobile converged network development is discussed. Section

III focuses on the possible heterogeneous scenarios in mobile

converged networks. Section IV covers coexistence solutions

and interference avoidance within converged heterogeneous

mobile networks. Section V concludes this paper.

II. OVERVIEW OF CONVERGED MOBILE NETWORK

DEVELOPMENT

A. First model and most recent model of converged mobile

networks

When developing the LTE standard, the 3rd Generation

Partnership Project (3GPP) introduced a new improved version

known as LTE-Advanced. LTE-A was the first step towards

network heterogeneity due to relay transmission possibilities.

LTE-A networks were not heterogeneous at first, because they

used only a single protocol and shared the same licensed

spectrum. However, the latest development allows for sharing

different unlicensed spectrum bands (Wi-Fi, Bluetooth, etc.) for

relay transmission [4] and supports effective handover between

LTE-A cellular networks and Wi-Fi or Bluetooth networks. In

particular, the convergence of two different network protocols,

Wi-Fi and LTE-A, is achieved via connection of fixed Wi-Fi

access points with LTE-A core networks through fixed or

wireless relay transmission. Mobile Wi-Fi/Bluetooth access

point technology was proposed to improve the flexibility of

converged mobile networks. This new converged technology

will also consider Machine-to-Machine (M2M)

communications sharing the available unlicensed spectrums of

Wi-Fi, Bluetooth, or ZigBee. An example of mobile converged

networks with different wireless access techniques is shown in

Fig. 1.

M2M network

Wi-Fi

networkCellular

network

M2M network

Cellular

users

Wi-Fi

usersWi-Fi

users

M2M

users Macrocell user/

M2M provider

Picocell user/

M2M provider

M2M network

Cellular user/

M2M provider

Cellular

users

Cellular

users

Cellular

users

M2M

users

M2M

users

Licensed spectrum of cellular network

Unlicensed spectrum of Wi-Fi network

Unlicensed spectrum of cellular network

Fig.1. Converged mobile network including LTE-A, Wi-Fi and M2M networks.

B. Three main types of converged mobile networks

We see three main types of converged mobile network:

device convergence, protocol convergence, and service

convergence. Device convergence was the first step, based on

integrating multiple communication interfaces within one

device. This type of convergence assumes the simultaneous use

of different types of separated communication sessions by one

device. As an example, a user is able to use an LTE-A network

for voice communication, simultaneously transferring data from

the smartphone to a laptop via Bluetooth. However, in this

convergence, each separate communication session is

controlled by the user. The second type of mobile networks is

protocol convergence. In the second type, different network

protocols (LTE, Wi-Fi, Bluetooth, etc.) are used within a single

heterogeneous network. As an example of this scenario, we


3

consider an LTE-A network with Wi-Fi access points, which

supports soft handover between cellular and Wi-Fi networks

controlled by the provider. In service convergence, continuous

mobile content service is the goal, regardless of device changes

and inter-network handover. Full convergence will be one type

where the other three types will be integrated. It will be

technically very complex, but an interesting future research

topic. The different convergence types and their features are

summarized in Table 1.

C. Future research topics in converged mobile networks

Full mobile network convergence will be a promising future

research topic model because it is technically very complex but

eventually provides total Quality of Service (QoS) to users. In

order to achieve full convergence in mobile networks, a lot of

technical problems must be solved. A wide range of new data

transmission methods over wireless channels must be

developed, such as energy-efficient modulation and coding

techniques, improved spectrum sensing, radio resource

management algorithms, etc. New wireless transmission

protocols for each network (LTE-A, Wi-Fi, M2M) must be

developed in order to achieve full interoperability between

them. The existing leading air interfaces based on Orthogonal

Frequency Division Multiple Access (OFDMA) and

Multiple-Input Multiple-Output (MIMO) antennas must be

improved to fit the demands of future converged networks.

There also must be a single signaling layer for the entire

converged network in order to maintain quality in the

experience, even for high-mobility users. M2M devices are

extensively used for metering, monitoring and control in

different areas, and the number of M2M devices is expected to

be huge. However, in order to be cost-effective, it is expected

that such M2M devices must be simple, consume little power,

and be reachable, robust and able to integrate with different

systems, such as TV, household appliances, alarm systems,

smart-house systems, cars, etc. It is possible that during mobile

network evolution towards 5G and 6G, the existing wireless

transmission protocols may be replaced by new advanced and

more efficient protocols.

III. WIRELESS HETEROGENEITY IN MOBILE CONVERGED

NETWORKS

A. M2M in mobile converged networks

M2M allows users to utilize multiple communication

interfaces while sharing network access with other users. M2M

networks usually use Wi-Fi or Bluetooth network protocols.

However, Wi-Fi and Bluetooth protocols provide quite a lot less

transmission capacity than cellular network protocols like

LTE-A and Worldwide Interoperability for Microwave Access

(WiMax). Due to this limited capacity problem with M2M

network protocols, the M2M and cellular convergence network

will be one of the prospective hot research topics in the future.

New spectrum reforming methods and algorithms were

proposed due to utilizing the unlicensed spectrum for cellular

network [5]. As further development of M2M and cellular

convergence networks, devices in the M2M network can share

the unlicensed spectrum of cellular networks like LTE-A [6].

Thus, M2M can reduce the transmission load and significantly

increase performance [7]. The four different possible scenarios

for M2M and cellular network convergence are presented in

Fig. 2.

TABLE I

COMPARISON OF FOUR DIFFERENT TYPES OF CONVERGENCE.

Type of

convergence

Objectives Subjects Features Stage of

development

Device

convergence

(Type 1)

Integration of different

communication

interfaces within single

end-use devices

Users’ devices and their

interfaces

Simultaneous holding

of separate

user-controlled

communication sessions

by a single device

Completely

achieved

Protocol

convergence

(Type 2)

Convergence of

different

communication

protocols within one

logical network

Communication

protocols

Soft handover between

different

communication

protocols without

session interruption

Almost

completely

achieved

Service

convergence

(Type 3)

Convergence of

network services under

one heterogeneous

mobile network

Network services and

users’ devices

Continuous service

maintained regardless

of user device and

connection type

Partially

achieved

Full

convergence

(Future)

Full convergence of any

user’s devices,

communication

protocols and network

services

Network services and

users’ infrastructures

Independence from

users’ devices,

communication

networks, and service

providers

None


4

M2M

network

M2M network

M2M

network

M2M

network

M2M network

M2M network

M2M

network

a) b)

c)d)

Cellular

usersCellular

users

M2M

users M2M

users

M2M

users

Multihop M2M users

Multihop M2M

users

M2M

users

Cellular user/

M2M provider

Cellular user/

M2M provider

Cellular user/

M2M provider Cellular users/

M2M providers

M2M

users

Cellular user/

M2M provider

Cellular user/

M2M providerCellular

network

Cellular

network

Cellular

network

Cellular

network

M2M

network

Multihop M2M

users

Cellular

users Cellular

users

Cellular users/

M2M providers

ΔF1 ΔF2 ΔF3

Whole spectrum

ΔF4

Fig.2. Converged mobile network including LTE-A, Wi-Fi and M2M networks.

TABLE II

COMPARISONS OF FOUR DIFFERENT M2M/CELL CONVERGENCE NETWORK SCENARIOS.

M2M/cell

network

convergence

scenario

Advantages Disadvantages

Hot spot–based

M2M network • simple deployment

• simple session control

• dynamic channel utilization

• possibility of overloading

• absence of quality control

mechanisms

• low reliability

Mesh-based

M2M network • traffic balancing

• multi-hop routes

• possibility of simple quality

control

• only static routes

• low reliability

• low channel utilization

• quality control mechanisms

Hybrid M2M

network • dense traffic balancing

• multi-hop routes

• dynamic channel utilization

• difficult deployment

• relatively low reliability

• only static routes

Fully connected

M2M network • multipath traffic control

• precise quality control

• dynamic multi-hop routing

algorithms

• high reliability

• difficult deployment

• complex session control

• complex routing control

• interference problem


5

The first and simplest approach is the hot spot–based M2M

and cellular convergence network in Fig. 2 a). In this scenario,

an M2M master node is selected from the criteria of maximum

performance and battery level. The master device will establish

the M2M area to be used to provide network access to other

devices. Thus, in this type of M2M/cellular convergence

network, the master M2M node utilizes all cellular resources,

and shares them with other M2M devices. However, this

scenario does not support any QoS control mechanisms. Thus,

in this case, the probability of M2M network overload is

relatively high if many devices are simultaneously connected to

the master device. For quality support reasons, the mesh-based

M2M and cellular convergence network scenario will be

preferred, because the convergence type uses the multi-hop

connections in Fig. 2 b). In this scenario, the master device

connects only a limited number of devices. Furthermore, each

device can be a secondary master, which can provide network

multi-hop access devices. This mode allows for the balancing of

traffic between users by managing data flows in the master

M2M device. In this type of converged network, the M2M

master device gets all the resources from the base station of the

cellular network and shares them, particularly between M2M

devices, through the unlicensed spectrum of cellular networks

like LTE-A. The third scenario is the hybrid M2M and cellular

convergence network, which can be organized by mesh

topology as shown in Fig. 2 c). This scenario supports hot

spot–based M2M sub-networks with master node, which is

connected by multi-hop communication channel. In general,

scenario in Fig.2 c) is combined from scenarios in Fig.2 a) and

Fig.2. b. This type of cellular/M2M convergence assumes that

data flows may be distributed in accordance to the users’ type of

service. The most advanced scenario is a fully connected M2M

network, which allows the use of multiple M2M connections for

one data flow. With this feature for each transmission, optimal

paths will be selected to achieve full unlicensed (Wi-Fi or

Bluetooth) spectrum utilization. Moreover, this scenario is

significantly different from previous ones, because all M2M

devices may simultaneously be cellular consumers and M2M

providers, thus increasing session reliability. Comparisons of

the four possible M2M and cellular convergence network

scenarios are summarized in Table 2.

B. Future research topics on M2M technology

Future M2M networks will apparently be based on a fully

connected topology, which can achieve the best performance.

There are three important problems for future research topics,

which must be solved to fully utilize the advantages of this

approach. First, the performance of end-user devices must

increase in order to support multiple simultaneous connections.

Performance increasing assumes designing the new end-user

devices architecture. The new architecture will be able to

support ultra-wide frequency range, high throughput M2M

transmission methods, multiple antenna arrays, and great

computing capabilities. Moreover, current network protocol

stack is not able to support interconnection between cellular

network and M2M underlay network. Therefore, converged

heterogeneous networks require a new protocol stack and a

multiservice service oriented platform in order to support

emerging services, such as ultra-high definition video

streaming, low latency real time applications, and mobile cloud

computing services. Important task for heterogeneous mobile

networks is service quality assurance. Therefore, it is necessary

to develop new algorithms for traffic management in mobile

networks with different services and effective utilization of

network resources. In [8], the advanced IP multimedia

subsystem (IMS-A) architecture was proposed for converged

heterogeneous mobile networks, which allows QoS provision

with service differentiation. The further research topic in this

field is to design new models for traffic balancing between

M2M and cellular links, and is to implement the quality of

experience (QoE) control policies. Second, the new algorithms

should be designed for converged mobile network’s control

plane. A promising solution for mobile network management is

SoftRAN (Software Defined Radio Access Network) [9].

SoftRAN is software defined centralized control plane for radio

access networks that abstracts all base stations in a local

geographical area as a virtual base station. The virtual base

station is comprised of a central controller, physical base

station, and master devices of M2M networks. SoftRAN

architecture provides load balancing, interference management,

as well as throughput and traffic control. The third but the most

important problem to address is interference problem.

Interference limits network performance greatly. Therefore, the

majority of future research into mobile convergence networks

should be devoted to interference mitigation and the coexistence

of different wireless standards, which is addressed in the next

section.

IV. INTERFERENCE AVOIDANCE PROBLEMS IN MOBILE

CONVERGED NETWORKS

A. M2M and cellular network coexistence within the same

spectrum

Establishing mobile M2M networks can achieve high spectral

efficiency by unlicensed spectrum utilization and efficient radio

resource management [10]. However, users’ mobility and M2M

network boundary uncertainty leads to a complex interference

problem between different M2M networks and between M2M

and cellular networks [11]. On the other hand, using different

spectrum pools for separate M2M networks decreases the total

capacity of a mobile converged network. As shown in Fig. 3, the

most urgent problem in the M2M/cellular converged network is

coexistence between cellular and M2M transmissions within

one spectrum.

A simple and straightforward solution is for the transmission

power of M2M devices to decrease within the M2M network.

This will cause significant interference in neighboring cells

when reusing the same frequency. On the other hand, decreasing

cellular transmission power will lead to significant imbalance

between uplink and downlink coverage. Using the existing

solutions for interference mitigation in static LTE-A networks is

not optimal due to different interference levels, compared to a

heterogeneous mobile network. Thus, other solutions will be

needed and will be addressed in the next section.


6

Fig.3. Spectrum reuse and interference within a converged mobile network of M2M and cellular networks.

B. Coordinated multipoint for interference avoidance

Co-channel interference management is of fundamental

importance to allow direct transmissions between devices

reusing radio spectrum of cellular networks. Coordinated

Multi-Point (CoMP) is a promising solution able to mitigate

and/or avoid high co-channel interference [12]. In fact, CoMP

schemes will play an important role of allowing the coexistence

of converged networks, M2M communications and cellular

networks within the same spectrum. CoMP will offer potentially

high benefits for future wireless networks in terms of spectrum

utilization and cellular coverage.

In CoMP systems, multiple transmission points are equipped

with one or more co-located antennas interconnected by a fast

backhaul (X2 interface for LTE-A). The fast backhaul allows

multiple cells to coordinate their transmissions in order to

improve signal strength and/or reduce inter-cell interference.

The most common CoMP transmission schemes include

coordinated scheduling/coordinated beamforming, and joint

processing with joint transmission.

Coordinated scheduling reduces inter-cell interference

among co-channel cells by scheduling users or silencing users.

In other words, one cell can decide to assign a channel to a

different user (scheduling) or even keep the channel unused

(silencing) in order to reduce interference with neighboring

cells. In a conventional system, the beamforming vectors in

each cell are set independently. But in the coordinated

beamforming system, the beamforming vectors for different

base-stations are coordinated to aim at minimizing interference.

For coordinated scheduling/coordinated beamforming, data of

all co-channel users are not needed for all cooperating cells (or

for a central controller (i.e., a fast backhaul)). They require only

a little information on scheduling decisions and channel quality

of the users [13].

When a joint transmission is made, data and Channel State

Information (CSI) of co-channel users are synchronously

available to all cooperating cells (or to a central controller) and

thus the transmit antennas work in a large multi-antenna system.

The transmit antennas can jointly process signals using the

well-known multi-antenna transmission schemes. One of the

well-known multi-antenna transmission schemes will be linear

precoding. The linear precoding is to coherently combine

signals of interest as well as to suppress interfering signals

coming from different cells.

C. Distributed Input Distributed Output Wireless

Technology (DIDO)

Distributed Input Distributed Output (DIDO) wireless

technology is a new revolution approach for multiuser data

transmission in wireless networks. DIDO allows utilization of

entire spectrum by each user [14], despite of sharing the same

spectrum between plenty users. DIDO provides each wireless

user with full data rate (bits/seconds) of shared spectrum

simultaneously with all other users by removing interference

between users sharing the same spectrum. In traditional wireless

technologies, the data rate available per user drops as more

users share the same spectrum to avoid interference, but in

DIDO, the data rate per user keeps the full data rate of the

spectrum as more users are added. DIDO uses the cloud DIDO

data center. The cloud DIDO data center controls all

transmission sessions within wireless network. In DIDO


7

systems instead of interference avoidance, all wireless

transmitters create interference artificially. Moreover,

interference is a key part of DIDO networks, because all

transmitted signals are summed in each receiver in clearly

modulated waveform carrying the intended data for each device.

First, DIDO transmitters exchange pilot test signals with the

DIDO receivers. By analyzing result of these test signals

propagation, the DIDO data center precisely determines the

channel conditions and calculates how the simultaneously

transmitted signals will sum together in each receiver. Then, the

DIDO data center creates precise waveforms for all

transmitters, and stimulates interference between these signals.

All signals will sum together at each user device creating a clean

independent waveform carrying the data requested by that user.

D. Future research topics of CoMP in mobile converged

networks

For future converged networks, a higher number of cells

(macro-, micro-, pico- and femtocells) are expected. Aside from

their intrinsic advantages, CoMP schemes could be further used

for converging networks to coordinate transmission on different

scales, such as coordinating scheduling and/or orthogonalizing

resources between different cell layers (e.g., macro- and

femtocells). They can be achieved by clustering transmission

points within a cell layer for joint processing or by coordinating

different technology underlays between M2M and conventional

cellular users. Most CoMP techniques consider a single

technology, but in converged networks, multiple technologies

(LTE, Wi-Fi, etc.) will be used. Different CoMP uses within

mobile converged networks are illustrated in Fig.4. CoMP

transmission schemes such as coordinated scheduling or

coordinated beamforming, and joint processing may be

employed in a cellular network in the future as illustrated in Fig.

4 a). Fig. 4 b) presents the kind of future CoMP scenario where

users are able to simultaneously utilize cellular and M2M

connections. This type of CoMP may be realized in different

ways. As an example, uplink transmission may utilize cellular

connection while downlink uses M2M, and vice versa. Another

scenario might use time division or frequency division

duplexing for these purposes.

A further possible evolution of CoMP is simultaneous

utilization of a few M2M connections shown in Fig. 4 c). This

approach needs a precise data flow control protocol, and

therefore will be useful only for fully connected M2M networks.

Full convergence can be achieved with CoMP between all

possible connections shown in Fig. 4 d). This type of CoMP is

of great interest, because it will be able to combine LTE-A

performance with WSN scalability, which perhaps will be a big

step towards 5G mobile networks.

Another hot research topic is the DIDO wireless networks.

DIDO has been already tested at frequencies from 1 MHz to 1

GHz. The test results have shown extremely great performance.

However, it seems obvious that increasing the number of DIDO

users will sufficiently increase the complexity of separate

waveforms calculation. Therefore, we assume that many future

researchers will be devoted to designing new computing

algorithms and signal modulation techniques for DIDO

networks.

M2M

eNB

eNB

eNB

M2MM2M

M2M

M2M

eNB

eNB

eNB

eNB

a) b)

c) d)

ΔF1 ΔF2 ΔF3

Whole spectrum

ΔF4 ΔF5

Fig.4. Possible uses of CoMP in mobile converged networks: a) conventional CoMP between cells, b) CoMP between cells and M2M networks,

c) CoMP between M2M networks, and d) CoMP between multiple cells and multiple M2M networks.


8

E. Case study: M2M communications underlaying cellular

networks by CoMP

Our case study illustrates potential spectrum utilization and

coverage gains achieved by CoMP transmissions in the

converged network of cellular and M2M communications

sharing the same spectrum. The system-level simulation was

conducted for a downlink OFDMA multi-cell system with

urban-microcell propagation environment, which is aligned

with 3GPP LTE architecture. Further models and simulation

parameters are detailed in [15].

To obtain resource reuse gains and exploit M2M proximity,

M2M communications are only enabled in hotspot zones near

cell-edge, while minimizing the impact on the performance of

cellular communications [15]. For joint processing, we consider

Maximum Ratio Transmission (MRT) in the desired link with

ideal channel state information. After resource assignment is

carried out for cellular communications by the Proportional Fair

(PF) scheduling policy, M2M devices are chosen based on the

degree of cooperation. Two degrees of cooperation are

considered in this simulation: CoMP between cells and M2M

networks, and CoMP between M2M networks, respectively

shown in Figs. 4 b) and c), and they are detailed as follows:

CoMP between cellular and M2M connections are

considered when a cellular user is scheduled inside the hotspot

zone. Hence, an M2M device is randomly selected inside the

hotspot for establishing a CoMP transmission.

The conventional scenario, in which M2M communications

are not enabled, is considered as baseline;

CoMP between M2M connections is considered in every

resource independently of the position of the cellular user. In

this case study, two users with the same data to transmit and one

receiver user are randomly selected inside the hotspot. As

baseline, the cellular user shares the same spectrum with a M2M

communication between two devices in each cell.

Fig. 5 shows the Cumulative Distribution Function (CDF) of

Signal to Interference-plus-Noise Ratio (SINR) levels of

cellular and M2M communications for both considered CoMP

schemes. As shown in Fig. 5 a), the performance is considerably

improved using another closely located device to enhance the

link quality perceived by cellular users. By comparing baseline

scenarios in Figs. 5 a) and b), notice that there is a high negative

impact on the performance of cellular communications due to

M2M communications. Many cellular users are out of coverage

with SINR levels below -6.2 dB, i.e., the minimum SINR target

to communicate [15] due to excessive interference generated by

M2M communications. However, the performance of both

M2M and cellular communications is improved when CoMP

transmission is employed, as shown in Fig. 5 b). While the gain

with M2M communications comes from joint processing, the

cellular gain happens because the energy conveyed by M2M

transmitters is directed to the receiver and so the interference is

weakened. Thus, CoMP transmission has been proven as a

promising solution to protect cellular communications.

Spectral efficiency gains for CoMP communications in

comparison to baselines of the CoMP scheme between cells and

M2M networks are 75%. Spectral efficiency gains for CoMP

communications in comparison to baselines of the CoMP

scheme between M2M networks are 29% (It means that 83% of

gains are for cellular networks). High spectral efficiency gains

are achieved. However, it is important to highlight that these

performance improvement comes at the cost of an increased

signaling and power consumption since more transmitters are

involved in the CoMP communication.

For the future communication scenarios we consider CoMP

as promising approach for improving the overall network

performance. Combining the CoMP, M2M and DIDO allows to

design the new fully converged architecture of wireless access

network, where each transmission link will be established by

using the optimal way for current network circumstances in

order to achieve the best energy and spectral efficiency.

Moreover, the software defined approaches can be

implemented for the convergent mobile network, providing the

very flexible and convenient methods for 5G RAN management

and service maintenance.

Fig. 5. SINR levels of cellular and M2M communications for two CoMP schemes: a) CoMP between cells and M2M networks, and conventional

scenario (No CoMP), and b) CoMP between M2M networks, and M2M communications underlaying cellular networks (No CoMP).


9

V. CONCLUSION

The evolution of current wireless mobile standards towards

the next generation will require a soft convergence of

heterogeneous mobile networks in order to provide the users a

unique and reliable mobile connection experience in different

types or services regardless of their location, connection type,

and device. This paper has surveyed the key challenges and

candidate solutions for the convergence of current wireless

standards towards a heterogeneous mobile network scenario.

Key problems related to the coexistence of the different wireless

standards (cellular, broadcast, Wi-Fi and Bluetooth) within the

same spectrum band have been discussed. The integration of

CoMP and M2M in a converged heterogeneous network entails

a number of research challenges ranging from cell identification

to physical and MAC layer problems such as interference

coordination and scheduling. These issues open new

perspectives for research in 5G wireless communications.

Through our case study, we have seen that CoMP transmissions

in heterogeneous networks, cellular and M2M networks offer

high gains in terms of spectrum efficiency and coverage for

mobile converged networks. Further spectral efficiency and

QoS improvements might be achieved allowing for an

intelligent (re)use of radio resources. Developing the DIDO

Networks is of great interest in future research for 5G. The

effective use of interference for distributed signal transmission

will be a great solution for future 5G wireless networks

ACKNOWLEDGMENT

This research was supported by the National Research

Foundation, the Korean government (MEST) (No.

2012-0004811).

REFERENCES

[1] M. N. Dumbre, M. M. Patwa and M. K. Patwa, “5G WIRELESS

TECHNOLOGIES-Still 4G auction not over, but time to start talking

5G”, International Journal of Science, Engineering and Technology

Research, vol.2, no.2, 2013, pp. 435-440.

[2] D. Sabella, P. Rost, Y. Sheng, E. Pateromichelakis, U. Salim, P.

Guitton-Ouhamou and G. Giuliani, “RAN as a Service: Challenges of

designing a flexible RAN architecture in a cloud-based heterogeneous

mobile network”, In Proc. of IEEE Future Network and Mobile Summit,

July 2013, pp. 1-8.

[3] M. Jo, L. Han, D. Kim and H.P. In, “Selfish Attacks and Detection in

Cognitive Radio Ad-Hoc Networks”, IEEE Network, vol.27, no.3, 2013,

pp. 46- 50.

[4] J. Cao and C. Zhang, “Seamless Roaming over Heterogeneous Wireless

Networks”, In Seamless and Secure Communications over

Heterogeneous Wireless Networks, Springer New York, 2014, pp. 27-62.

[5] P. Guskov, R. Kozlovskiy, T. Maksymyuk and M. Klymash, “Methods

and techniques of spectrum reforming for LTE network deployment”, In

Proc. IEEE International Conference on Microwave &

Telecommunication technology (CriMiCo’2013), September 2013, pp.

474-475.

[6] V. Osa, C. Herranz, J. F. Monserrat and X. Gelabert, “Implementing

opportunistic spectrum access in LTE-advanced”, EURASIP Journal on

Wireless Communications and Networking, no.1, 2012, pp. 1-17.

[7] Y. Chen and W. Wang, “Machine-to-Machine communication in

LTE-A”, In Proc. of 72nd IEEE Vehicular Technology Conference Fall

(VTC 2010-Fall), 2010, pp. 1-4.

[8] M. Klymash, B. Strykhalyuk, M. Beshley and T. Maksymyuk, “Research

and Development the Methods of Quality of Service Provision in Mobile

Cloud Systems”, In Proc. IEEE International Conference on

Communications and Networking (BlackSeaCom’2014), Odessa,

Ukraine, May 2014, pp. 165-169.

[9] A. Gudipati, D. Perry, L. E. Li and S. Katti. “SoftRAN: Software defined

radio access network”, In Proceedings of the second ACM SIGCOMM

workshop on Hot topics in software defined networking, August 2013,

pp. 25-30.

[10] C. F. M. Silva, J. M. B. Silva and T. F. Maciel, “Radio Resource

Management for Device-to-Device Communications in Long Term

Evolution Networks”, 1st ed., F. R. P. Cavalcanti, Ed. Springer New

York, 2014, http://dx.doi.org/10.1007/978-1-4614-8057-0 3.

[11] P. Janis, V. Koivunen, C. Ribeiro, K. Doppler and K. Hugl,”

Interference-avoiding MIMO schemes for device-to-device radio

underlaying cellular networks”. In Proc. of Personal, 20th IEEE

International Symposium on Indoor and Mobile Radio Communications,

2009, pp. 2385-2389.

[12] E. M. G. Stancanelli, R. L. Batista, T. F. Maciel and Y. C. B. Silva,

“Radio Resource Management for Coordinated Multipoint Systems”, 1st

ed., F. R. P. Cavalcanti, Ed. Springer New York, 2014. [Online].

Available: http://dx.doi.org/10.1007/978-1-4614-8057-0_1.

[13] J. Dai, M. Zhao, L. Jiang, M. Chen, Z. Ni and Y. Li, “Regional Planning

of Coordinated Multi-Point Transmission with Perfect Feedback”,

Wireless Personal Communications, New York: Springer New York, Feb.

2014, pp. 1-21.

[14] A. Forenza, R. W. Heath Jr, S. G. Perlman, R. Van der Laan and J. Speck,

“System and method for distributed input distributed output wireless

communications”, U.S. Patent No. 8,428,162. Washington, DC: U.S.

Patent and Trademark Office, 2013.

[15] R. L. Batista, C. F. M. Silva, J. M. B. Silva Jr., T. F. Maciel and F. R. P.

Cavalcanti. “What happens with a proportional fair cellular scheduling

when D2D communications underlay a cellular network?”, In Proc. of

IEEE WCNC 2014 - Workshop on Device-to-Device and Public Safety

Communications (WCNC'14 - WDPC Workshop), Istanbul, Apr. 2014,

pp. 260-265.

Minho Jo (M’07) is now Professor of Computer and

Information Science, Korea University. He received his

Ph.D in industrial and systems engineering, Lehigh

Univ., USA in 1994 and BA in industrial engineering

in Chosun Univ., S. Korea in 1984. He was one of

founding members and the Founding Captain of IT

Team of LCD Division, Samsung Electronics. He has

an extensive experience of wireless communications

and software development in industry for more than 15 years. He is now

Founding Editor-in-Chief of KSII Transactions on Internet and Information

Systems. He serves now as an Editor of IEEE Network, and Editor of IEEE

Wireless Communications. He is now Associate Editor of Security and

Communication Networks, and Associate Editor of Wireless Communications

and Mobile Computing, respectively. He is currently Vice President of Institute

of Electronics and Information Engineers (IEIE), and Vice President of Korea

Information Processing Society (KIPS), respectively. The topics of his current

interest of research include cognitive radio, mobile cloud computing, network

security, heterogeneous network in 5G, cellular network in 5G, and IoT

(Internet of Things).

Taras Maksymyuk (M’14) received his M.S. degree

in information communication networks, Lviv

Polytechnic National University, Ukraine in 2011, and

BA degree in telecommunications, Lviv Polytechnic

National University, Ukraine in 2010. Currently, he is

with the Department of Computer and Information

Science at Korea University, Sejong City, South Korea.

Current Member of IEEE Communications Society and

IEEE Cloud Computing Community. His research

interests include cognitive radio networks, converged networks, mobile cloud

computing, massive MIMO in 5G, machine-to-machine communication and

heterogeneous network in 5G, software defined radio access networks and

Internet of Things.


10

Rodrigo L. Batista received the B.Sc. degree in

Computer Engineering from Federal University of

Espírito Santo, Brazil, in 2008, and the M.Sc. degree in

Teleinformatics Engineering from the Federal

University of Ceará, Brazil, in 2011. Between 2009

and 2014 he worked as a Research Engineer with the

Wireless Telecommunications Research Group

(GTEL), Brazil, in projects within technical

cooperation with the Ericsson Research, Sweden. His

research interests include 5G wireless communication networks, inter-cell

interference management, coordinated scheduling, design aspects of

network-assisted device-to-device communications for opportunistic cellular

spectrum re-utilization.

Tarcisio Ferreira Maciel (S’01-M’09) received the

B.Sc. and M.Sc. degrees from the Federal University of

Ceará (UFC), Fortaleza, Brazil, in 2002 and 2004,

respectively, and the Dr.-Ing. degree from Technische

Universität Darmstadt (TUD), Darmstadt, Germany, in

2008, all in electrical engineering. From 2001 to 2004,

he was a researcher with the Wireless Telecom

Research Group (GTEL), UFC. From 2005 to 2008, he

was a research assistant with the Communications

Engineering Laboratory, TUD. Since 2008, he has been a researcher with

GTEL and a member of the Post-Graduation Program in Teleinformatics

Enginnering (PPGETI) of the UFC. In 2009, he was a Professor of computer

engineering with the UFC, and since 2010, he has been a Professor with the

Center of Technology, UFC. His research interests include radio resource

management, numerical optimization, and multiuser/multiantenna

communications.

André L.F. de Almeida (M’08–SM’13) received the

double Ph.D. degree in Sciences and Teleinformatics

Engineering from the University of Nice, Sophia

Antipolis, France, and the Federal University of Ceará,

Fortaleza, Brazil, in 2007. He is currently an Assistant

Professor with the Department of Teleinformatics

Engineering of the Federal University of Ceará. During

fall 2002, he was a visiting researcher at Ericsson

Research Labs, Stockholm, Sweden. From 2007 to

2008, he held a one year research position at the I3S Laboratory, CNRS,

France. In 2008, he was awarded a CAPES/COFECUB research fellowship

with the I3S Laboratory, CNRS, France. He currently holds a productivity

research fellowship from the Brazilian National Council for Scientific and

Technological Development (CNPq). In spring 2012, he was a visiting

professor at the University of Nice-Sophia Antipolis, France. He is currently

Associate Editor of the IEEE Transactions on Signal Processing, KSII

Transactions on Internet and Information Systems, and Circuits, Systems and

Signal Processing. His research interests include the topics of channel

estimation, equalization, array processing, collaborative networks, and

tensor-based methods with applications to communications and data analysis.

Mykhailo Klymash (M’13) is now the Chief of

Telecommunication Department, Lviv Polytechnic

National University. He received D.Sc. degree in

telecommunication systems and networks, Odessa

National Academy of Telecommunication named after

A.S. Popov, Ukraine in 2007, and his Ph.D in optical

data transmission, location and processing systems,

Bonch-Bruevich Saint-Petersburg State University of

Telecommunications, Saint Petersburg, Russia, in

1994. Current member of IEEE Communications

Society. Editor of Radioelectronics and Telecommunications Issue, Journal of

Lviv Polytechnic National University. Laureate of the Ukrainian State Award

in Science and Technology in 2012. Honored member of Ukrainian

Communications Academy. The topics of his current interest of research

include distributed networks, cloud computing, convergent mobile networks,

big data, software defined networks and 5G heterogeneous networks.

a survey of converging solutions for heterogeneous mobile...

Documents